Thesis: Pallet Perception with DL & Sensor Fusion

Toyota Material Handling Manufacturing Sweden; Mjölby kommun; On-site; SEK 78,000 - 112,000; Part time

Job summary

Toyota Material Handling Manufacturing Sweden invites applications for a Master thesis focused on developing methods using deep learning to improve pallet perception. The project emphasizes fusing model outputs with RGB/ToF data to localize pallets, in collaboration with Toyota's ML teams, starting January 2027.

The thesis spans 30 hp with 1–2 students, incorporating synthetic and real-world data, and leveraging an NVIDIA GPU in an embedded environment for data processing at Toyota's Mjölby site.

Qualifications

• 3D computer vision expertise required.
• Deep learning / ML specialization required.

Responsibilities

• Train and evaluate deep learning models using different methodologies.
• Develop algorithms for data fusion between sensors and model outputs.
• Evaluate trained models with real-world data and synthetic data, and compare results.

Skills

3D computer vision

Deep learning specialization

Job description

Toyota Material Handling Manufacturing Sweden invites applications for a Master thesis focused on developing methods using deep learning to improve pallet perception. The project emphasizes fusing model outputs with RGB/ToF data to localize pallets, in collaboration with Toyota's ML teams, starting January 2027.

The thesis spans 30 hp with 1–2 students, incorporating synthetic and real-world data, and leveraging an NVIDIA GPU in an embedded environment for data processing at Toyota's Mjölby site.